| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Biochemistry and Biophysics |
Magnetic Resonance Imaging of RIF-1 Tumors in Vivo
Departments of Radiology [U. D., H. P., L. N-D., R. R., J. S. L., J. D. G.], Pathology and Laboratory Medicine [M. F.], and Medicine [M. S. G., W. M. F. L.], University of Pennsylvania, Philadelphia, Pennsylvania 19104-6100
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
) and the transverse relaxation time (T2). Measurements of these relaxation parameters were performed on a s.c. murine radiation-induced fibrosarcoma (RIF-1) model before and after cyclophosphamide treatment. The number of pixels exhibiting T1 values longer than controls in viable regions of the tumor increased significantly as early as 18 h after drug administration and remained elevated up to 36 h after treatment (P < 0.005). Although a trend of increasing T2s relative to controls was noted in viable regions of the tumor 36 h after treatment, the changes were not statistically significant. Histological examination indicated a decrease in mitotic index that paralleled the changes in T1. We conclude that T1 measurements may be useful for noninvasive monitoring of early response of tumors to chemotherapy. |
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
MRI4 and spectroscopy have been used to noninvasively assess the pathophysiological status of tumors (6, 7, 8) . Magnetic resonance spectroscopy has proven to be an effective method that can evaluate the biochemical status of various metabolites (9) and to demarcate the early changes associated with cell death (10 , 11) . Diffusion-weighted spectroscopy can detect RIF-1 tumor response to cyclophosphamide therapy within 2 days (12) . However, the disadvantage of spectroscopic measurements is the loss of spatial information that can depict heterogeneity in the cellular response.
High-resolution imaging techniques can potentially detect response in discretely localized regions of the tumor. Therefore, it may be possible to monitor and evaluate the synergistic effects of multidrug therapies by visualizing the subpopulation of cells killed by the individual therapeutic modalities. Consequently, high-resolution imaging techniques may facilitate early-stage optimization of therapeutic regimens tailor fitted to the needs of individual patients. Several authors have used T2- and diffusion-weighted imaging to detect therapeutic response in animal models (13, 14, 15) within 2 days of treatment.
Of the various parameters that can be measured with MRI, T1 has been proposed to be sensitive to the macromolecular composition of tissues (16, 17, 18)
. Because such changes are anticipated in the early stages of tumor response to chemotherapy, T1 MRI may be an early indicator of therapeutic response. The present study was designed to determine whether T1 and/or T2 MRI could detect tumor response to an alkylating agent, cyclophosphamide, within 24 h.
Spatially resolved measurements of T1 and T2 were performed in a RIF-1 tumor model before and after treatment with cyclophosphamide. The MRI data were correlated with morphometric analyses of histological sections. Images were segmented to delineate viable regions of the tumor, to which the relaxation analysis was restricted. These data suggest that T1 measurements may be useful in detecting tumor response as early as 18 h after chemotherapy administration, whereas T2 changes are less sensitive in detecting response.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Animals and Tumors.
The University of Pennsylvania Institutional Animal Care and Use Committee approved all animal procedures. Subcutaneous tumors were induced in female C3H/HeN mice (18–20 g; Taconic Farms Inc., Germantown, NY) by inoculating 3 x 105 cells in the upper thigh. Tumor volumes were determined by caliper measurements of three orthogonal diameters (X, Y, Z), using the formula V = 
(X x Y x Z)/6. Tumor volumes were 175–300 µl in all experiments except growth delays (see below).
Cyclophosphamide Treatment.
Animals were treated with a single dose of cyclophosphamide dissolved in saline solution, which was injected i.p. The drug doses were varied (between 50 and 200 mg/kg) for the growth delay (n = 29) and cell survival experiments (n = 12). For all MRI experiments (n = 9), a dose of 200 mg/kg of body weight was administered.
Cell Survival Assays.
Tumor clonogenic assays were performed as described previously (19)
. Briefly, 24 h after the cyclophosphamide dose, tumors were excised, minced, suspended, and mixed in 10 ml of an enzyme cocktail. The enzyme digestion was stopped with 10 ml of serum-containing medium, and the suspension was filtered. The cells were counted and centrifuged at 4°C for 10 min at 1000 rpm. The pellet was then resuspended in medium to produce a suspension of 106 cells/ml. Cell dilutions were made, and appropriate numbers of cells were plated in 100-mm2 tissue culture plates. After incubation for 10–12 days, the colonies (
50 cells) formed were fixed, stained, and counted. Plating efficiency was calculated based on the number of cells plated, and cell survival was normalized to 1.0.
Growth Delay Studies.
Tumors were treated with cyclophosphamide at day 13 after tumor cell inoculation. The difference in the time required to reach a predetermined volume (1000 mm3) for control test groups defined the absolute growth delay.
Total Protein Content.
Tumor tissue was excised from the animals (n = 9) immediately after sacrifice. The tissue was stored in liquid nitrogen until further use. All samples were homogenized in ice-cold perchloric acid. The homogenate was then centrifuged at room temperature for 5 min at 12,000 rpm to obtain a protein pellet. The total protein content was estimated with a Lowry protein microassay kit (Sigma Chemical Co., St. Louis, MO). The measurements were normalized to the initial wet weight of the tumor.
MRI Measurements.
The animals were divided into two age-, tumor volume-, and weight-matched cohorts: saline-treated controls (n = 4) and animals treated with 200 mg/kg cyclophosphamide (n = 5). Animals were imaged at 0, 18, and 36 h after cyclophosphamide administration. The 0 time point was defined as the time when the tumor volume was 
200 µl. All experiments were performed on a General Electric Signa 4T clinical scanner. The animals were anesthetized with an i.p. injection of Ketamine (155 mg/kg) and Acepromazine (1.55 mg/kg). The animals were placed in a plastic holder, with the tumor protruding through a circular opening. External saline fiducials were placed next to the tumor to facilitate slice localization in successive imaging experiments. The position of the imaging plane was determined by measuring the distance from the external markers to the imaging slice of the preceding experiments. A four-turn, 4-cm diameter solenoid coil was used for all MRI experiments.
T2 measurements were made with a spin-echo sequence (single echo), using the following parameters: TR = 2 s; TE = 30, 50, 80, and 95 ms; slice thickness, 0.9 mm; interslice spacing, 0.2 mm; field-of-view, 4 cm; acquisition matrix, 256 x 128. The total imaging time for obtaining a T2 map was 17 min.
After the T2 measurements, a single slice through the center of the tumor was chosen for T1 measurements, which were performed with a T1-prepared fast spin echo sequence (20)
. The following imaging parameters were used: TR = 1 s; TE = 17 ms; slice thickness, 0.9 mm; field-of-view, 4 cm; and a 256 x 128 acquisition matrix. T1 values were measured by varying the spin-locking time (TSL = 10, 30, 50, 100, and 125 ms). Because T1 is known to vary with the strength of the spin-locking field, T1 was also measured at spin-lock powers of 0.117, 0.235, and 0.352 G. All three T1 relaxation maps were measured in 8 min. For detection of hemorrhage, a T1-weighted image (TR = 300 ms; TE = 7 ms) was obtained at the end of the experiment. The entire MRI examination required a total of 40 min.
Data Analysis.
The images were transferred to a Silicon Graphics computer and were analyzed with routines written in IDL (RSI, Boulder, CO). The spatial maps of T2 relaxation data were calculated by fitting the data on a pixel-by-pixel basis to Eq. A
. Similarly, T1 relaxation maps were obtained by fitting the data to Eq. B
, where TE is the echo time:
 |
 |
dispersion curve was determined from the ratio of the T1 maps obtained at 0.352 and 0.117 G. Any pixels that were fit with a correlation coefficient (r) <0.95 were set to zero and appeared black on the relaxation images. This was done to exclude pixels that did not fit a single exponential.
Image Segmentation and Analysis.
The relaxation maps were then segmented based on a region-of-interest analysis to demarcate viable and nonviable tumor. Edema and necrosis appear as hyperintense regions on a T2-weighted image (21)
. Similarly, acute hemorrhage appears dark on both T2- and T1-weighted sequences (22)
. We therefore avoided any areas that were not consistent with the signal characteristics of the bulk tumor. Nonviable regions were defined as regions that were hyper- or hypointense relative to the bulk tumor on relaxation maps and were morphologically well demarcated from the rest of the tumor. The viable areas of the tumor were segmented with a freehand drawing IDL routine (RSI). The means and SDs of the relaxation data from these regions were then determined.
A histogram analysis was performed to further characterize the distribution of the relaxation times in the viable regions of the tumor. Histograms were calculated using IDL routines, and the area under the histogram was normalized to unity. The normalization procedure compensates for differences in population size. The normalized histograms obtained from the various cohorts were then analyzed for trends.
The mean relaxation values from the control cohort were used to establish threshold values above which the response to cyclophosphamide was quantified. The threshold values were taken to be more than 2 SD above the mean values of the region of interest. The percentage of those pixels that were above the threshold was then determined for both cohorts. The threshold values were as follows: T2 = 60 ms; T1 (0.177G) = 70 ms; T1 (0.235 G) = 75 ms; T1 (0.352 G) = 80 ms.
Histological Analysis.
At the end of the MRI examination, the animals were sacrificed by an overdose of i.p. anesthesia and cervical dislocation. The tumors were then dissected along with the skin and s.c. tissues and placed in 10% buffered formalin. Tissue sections were then cut parallel to the MRI plane. Tissue sections were obtained at 5-µm thickness and stained with H&E. The mitotic indexes, as well as morphological measurements of the percentages of areas of gross cell death to total tumor area, were measured for each animal. Mitotic and apoptotic counts were obtained by morphological evaluation of 20 hpfs (x40) /animal. Areas of gross cell death were quantified by image morphometry by dividing the total area of tumor death by the total tumor area as measured using ImageTool software (developed at the University of Texas Health Science Center San Antonio).
Formalin-fixed, paraffin-embedded tumor sections (4-µm thickness) were stained for apoptosis by the TUNEL method (Apoptag kit; Intergen, Purchase, NY), according to the manufacturer’s protocol. Digoxigenin-conjugated nucleotides were detected using FITC-conjugated antidigoxigenin antibody provided in the kit. Apoptotic nuclei were counted by a fluorescent microscope at x10 magnification. At least three fields were counted per section.
Statistical Analysis.
All data are reported as mean ± SE. Analysis for statistical significance of difference between two groups was performed with a Student’s t test. ANOVA was used to test for differences between the various time points for the cell survival assay. Repeated-measure ANOVA was used to determine statistical significance for the MRI measurements. Linear regression analysis with clustering was used to determine the correlation between MRI changes and treatment. ANOVA and t test analyses were performed using the SigmaStat program (SPSS, Chicago, IL); regression analysis was performed with Stata software (Stata Corp., College Station, TX). Any changes in the histogram data were analyzed for statistical significance by a one-way ANOVA based on ranks on the SigmaStat program. P < 0.05 was considered significant.
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. However, the retardation in tumor growth caused by drug administration (>50 mg/kg) became more obvious at longer time points (e.g., day 4 after treatment). After day 5, the cohort of animals treated with 200 mg/kg cyclophosphamide had a significantly smaller tumor volume than control animals (P = 0.046). The tumors in the controls and the cohort treated with 50 mg/kg grew to a volume of 1000 mm3 in 6 days. The growth delays for the cohorts treated with 100 and 200 mg/kg were 9 days and >9 days, respectively.
|
. The cohort of animals treated with 200 mg/kg exhibited the smallest cell survival fraction. The difference in cell survival fraction between the treatment groups was highly significant (P < 0.001, ANOVA). Therefore, all additional experiments were performed with a cyclophosphamide dose of 200 mg/kg.
Evidence That T1 Changes Detect Tumor Response in Viable Areas of the Tumor.
The percentage of pixels having long T1 values (at 0.177 G) did not change during the course of the study in the control animals (P = 0.3, ANOVA; Fig. 2A
). The percentage of pixels with a prolonged T1 (at 0.117 G) was significantly different in the treated animals compared with the controls. The percentage of pixels exhibiting prolonged T1 increased significantly 18 and 36 h after cyclophosphamide administration (P = 0.006, ANOVA). The percentage of pixels above threshold at 18 h increased by 45% compared with the initial measurements on the same cohort of animals (P < 0.001). Furthermore, the percentage of pixels above threshold 36 h posttreatment increased by 64% compared with the initial measurements (at time 0 h) on the same cohort of animals (P < 0.001). The increase in the number of pixels with long T1 values correlated with the time elapsed after drug administration (R2 = 0.62; P = 0.036).
|
maps obtained with spin-lock fields of 0.235 and 0.352 G did not yield significant differences in the number of pixels greater than threshold between treated and control groups (P = 0.097 and 0.5, respectively). However, the histopathology of the tumors demonstrated significant changes in the mitotic indices over the course of the study.
Changes in T2 following cyclophosphamide administration exhibited trends different from the trends in T1 at 0.117G. The percentage of pixels that were larger than the T2 threshold value did not increase in the control group of animals during the course of the study (P = 0.76, ANOVA; Fig. 2B
). There was no significant increase in the number of pixels having an elevated T2 value in the treated cohort of animals as a function of time after treatment (P = 0.93, ANOVA). In contrast to the T1 data, the percentage of pixels having long T2 relaxation times failed to correlate with time after treatment (R2 = 0.0056; P = 0.625). The percentage of pixels above threshold was largest at 36 h after treatment. However, this increase was not statistically significant when compared with either the control cohort or to the initial measurements on the same group of animals (P = 0.61 and 0.73, respectively).
The effects of cyclophosphamide therapy on the T1 maps in a representative animal are shown in Fig. 3
. Pixels with T1 > 80 ms are depicted in color in the second row of images. These pixels appear to be heterogeneously distributed throughout the tumor. The average normalized histogram (from all four treated animals) showed that the number of pixels with long T1 values increased after cyclophosphamide treatment (P < 0.001, ANOVA). The number of pixels clustered at the mean value was slightly decreased. These data suggest that some fraction of pixels demonstrate increased relaxation times after treatment. However, the histograms obtained from the control cohort did not significantly change during the course of the experiment (P = 0.259, ANOVA).
|
-weighted image, a T1 map for a treated animal (at 0.117 G), and the distribution of relaxation times for the nonviable regions of tumors are shown in Fig. 4
. Both the T1-weighted image and the map demonstrated a hyperintense region, which was segmented (curvilinear white line) and excluded from further analysis. No significant changes in the distribution of relaxation times for the nonviable regions were noted during the course of the experiment (P = 0.067, ANOVA). The average fraction of pixels excluded from the histogram analysis based on T1 maps from the treated cohort (because they originated from regions of the tumor that appeared to be necrotic/hemorrhagic on the basis of criteria described previously) was 8.5 ± 3% at 0 h, 17 ± 9% at 18 h, and 19 ± 1% at 36 h. These percentages were not significantly different from each other (P = 0.465, ANOVA). The fraction of pixels discarded from the control cohort (for the same reason) was 24 ± 3% at 0 h, 15 ± 1% at 18 h, and 24 ± 2 at 36 h; these percentages were statistically not different from each other (P = 0.102). Some tumors appeared homogeneous on the relaxation maps and were not segmented.
|
-weighted MRIs and the corresponding histological sections obtained from an animal whose tumor exhibited a large increase in necrotic fraction (based on histology and imaging) after cyclophosphamide treatment are shown in Fig. 5
; data from this animal were not used in the relaxation analysis. A large hyperintense region is clearly visible in both T1- (Fig. 5, C and D)
and T2-weighted (Fig. 5B)
images obtained 36 h after cyclophosphamide therapy. The histological section (Fig. 5E)
verifies that the bright region is eosinophilic and consists mostly of cell debris. Note that the demarcation of the hyperintense region is comparable in both the T1- and the T2-weighted images.
|
. Control animals showed a very high mitotic count along with areas of spontaneous cell death. In contrast, animals treated with cyclophosphamide showed a low mitotic count with similar areas of cell death.
|
TUNEL staining demonstrated that apoptotic cell death was induced by cyclophosphamide (Fig. 6, f–h)
. There was a trend toward increased apoptosis as a function of time after treatment. Apoptosis counts were not obtained on each animal for every time point.
The total protein content of treated tumors (18 h after drug administration) was 25% less than that of control tumors (n = 3). These data are in agreement with published reports indicating that cyclophosphamide has an inhibitory effect on protein synthesis (23) . The Lowry assay provides information about the protein content of the entire tumor sample; therefore, the total protein content could be directly correlated with the spatially resolved MRI data. Nonetheless, these data suggest that cyclophosphamide treatment may alter the protein content of the tumor.
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
and/or T2 could detect changes in tumor response to chemotherapy. Our data show that the T1 characteristics of tumors are altered by chemotherapy before changes in tumor volume are evident. These changes persist up to 36 h after drug administration. Significant changes in T2 were not observed during the same time period.
T2- and diffusion-weighted MRI techniques have been used to monitor tumor response to chemotherapy (12, 13, 14)
. The fractional water content of the extracellular space influences both the apparent diffusion coefficient and T2. Diffusion-weighted spectroscopy has been used to detect changes 2 days after cyclophosphamide administration in the RIF-1 tumor model. Significant changes were not observed within the first day, presumably because the free water content of the extracellular space was not significantly altered during this time (12)
. In the present study, the absence of significant elevations in T2 up to 36 h after drug administration suggests that the free water content of the extracellular space was probably not altered during the course of our experiment. Galons et al. (15)
found changes in the apparent diffusion coefficient of water within 2 days after treatment of MCF-7 tumors with paclitaxel and suggested that diffusion measurements might be the most sensitive method to detect early treatment response. Comparison of the sensitivities of T1 and diffusion imaging as indices of therapeutic response will be the subject of a future investigation.
Braunschweiger (24) reported that the total extracellular water content of RIF-1 tumors increases by 60% two days after treatment with 150 mg/kg cyclophosphamide. This would be expected to change T2 relaxation times. However, no significant changes in T2 were observed during the course of our experiment. Tumors were monitored for up to 36 h after treatment with 200 mg/kg cyclophosphamide. It may be difficult to compare these data with Braunschweiger’s observations (24) in light of the disparity between drug dosage and timing of measurements. There may also be some genetic drift in the RIF-1 tumor line that contributes to these differences. In both the control and treated groups, the areas of spontaneous cell death exhibited features of necrosis. Surprisingly, the RIF-1 tumors studied demonstrated well-defined areas of cell death compared with studies in the literature (12 , 19 , 24) . This is most probably attributable to genetic drift in the cell line. Various lines of evidence confirming genetic drift in the RIF-1 line (including hypoxic fraction measurement, EF5 binding, and Eppendorf electrode measurements) will be presented elsewhere. The purpose of this work was to present a novel MRI technique for evaluating tumor response. The presence of these areas of cell death does not affect our interpretation of the results because these areas were segmented out prior to analysis.
Diffusion-weighted studies have also been shown to be useful in detecting tumor response of brain tumors after viral therapy (10
, 11
, 13
, 14)
. Poptani et al. (13)
showed that apparent diffusion constant measurements can detect spatially localized tumor response to gene therapy 3 days after treatment, although the macroscopic measurements did not show a significant change until 9 days posttreatment. Similarly, we have found that T1 is altered in a spatially heterogeneous manner, although the mean of the T1 histograms did not change substantially (Figs. 3
and 4
). Histogram analysis (Fig. 3g)
showed that there were more pixels with elevated T1 values after treatment. In addition, the number of pixels clustered about the mean (i.e., the amplitude of the normalized histogram) was slightly but significantly (P < 0.001) reduced in a time-dependent manner after treatment. These observations show that cyclophosphamide therapy led to an increase in T1 of certain pixels in the tumor. Although extreme care was taken to sample the same slice on subsequent measurements, the dynamics of tumor growth in control animals and tumor shrinkage in treated animals made it impossible to ascertain that precisely the same imaging slice was measured on successive examinations. This could lead to some sampling errors in the present study. However, the inclusion of a cohort of animals allowed us to evaluate our data in the context of the sampling errors. The fact that the changes in T1 agree with growth delay measurements demonstrates that despite these inherent sampling errors, T1 measurements can be used as early predictors of response. Three-dimensional imaging of the entire tumor can be used to minimize sampling errors; however, this would require hours of data accumulation, which is impractical experimentally and clinically. Therefore, at the present state of the art, the best that can be done is to try to monitor the slices in the same region on successive measurements.
This study demonstrates that T1 values were altered within 18 h after drug administration in tumors that exhibited growth delays and decreased clonogenic survival. The selective increase in the percentage of pixels having prolonged T1 may be attributable to various mechanisms, including (a) changes in the water content of the extracellular space (17)
, (b) diffusion through inhomogeneous fields (25)
, (c) chemical exchange of protons (26)
, and/or (d) the motion of large macromolecules (27)
. Alteration of the macromolecular composition of the tumor induced by cyclophosphamide treatment may have led to the observed increase in the number of pixels with prolonged T1, possibly through the aforementioned mechanisms. This explanation is supported by the apparent correlation between the regions of prolonged T1 and the heterogeneous hematoxylin staining. However, the large difference in the spatial resolution of the MRI and histological techniques do not permit a direct comparison between the two techniques.
The potential for using T1 relaxation to study macromolecular dynamics has been described in the literature (16
, 28
, 29)
. However, we are not aware of any reports validating this technique in vivo. Cyclophosphamide is known to reduce protein synthesis by alkylating DNA, leading to cross-linking and strand breakage (23)
, which should lead to changes in DNA expression (i.e., protein content). There may also be changes in protein content caused by alkylation after cyclophosphamide treatment. Our measurements of total protein content indicate that cyclophosphamide therapy may be altering the protein content of the tumor within 18 h. Therefore, our data suggest that T1 relaxation may be sensitive to changes in macromolecular content induced by chemotherapy. Because cyclophosphamide has other effects in addition to disruption of protein synthesis, T1 measurements need not be uniquely sensitive to protein content. Cyclophosphamide has antivascular effects (24
, 30)
; the spatial heterogeneity in tumor response might be related to these effects. This explanation is consistent with the time course of the observed changes, as well as the spatially heterogeneous distribution of the responding pixels (Fig. 3)
.
In the present study, a spin-lock strength of 0.117 G (500 Hz) was found to yield the most consistent results. It remains unclear as to why the T1s measured at a spin-lock strength of 0.117 G (500 Hz) yielded much more consistent results than those measured at 0.235 G (1.0 KHz) or 0.352 G (1.5 KHz). The TUNEL staining demonstrated that cell death occurred at least partially through apoptosis after cyclophosphamide treatment. Therefore, we propose that this disparity may be the result of disruption of macromolecular aggregates that occurs during apoptotic cell death (31)
. Because the fraction of DNA in the overall macromolecular pool in the cell is small, it is unlikely that breakdown of DNA will significantly alter the hydrodynamic properties of the cell that modify T1. However, we postulate that changes in the protein and RNA content of the cell during apoptosis might contribute to such changes in T1, as observed by the differences in the total protein content of the treated versus control tumors. Because this was a longitudinal study and the animals were imaged before and after treatment, we were not able to obtain apoptotic counts on every animal for each time point. We therefore can comment only that apoptotic cell death pathways are active after cyclophosphamide therapy; we cannot, at present, quantify the time dependence of the apoptotic response.
One of the animals in our study exhibited a dramatic necrotic response after cyclophosphamide administration, based on histological sections. These areas were also visible in the T1- and T2-weighted images (Fig. 5)
. T1 measurements are not likely to yield any additional information in such cases. However, T1 measurements appear to be useful in detecting subtle changes in viable areas of the tumor.
Our data indicate that elevations in T1 are associated with tumor response to chemotherapy. The comparison of the relaxation time changes induced by chemotherapy administration in drug-resistant versus drug-sensitive tumors would help to directly establish the correlation between elevations in T1 relaxation and the cellular response to drug administration. Unfortunately, it often is not possible to directly compare drug-resistant and drug-sensitive tumors because these tumors are phenotypically different. However, by performing dose-response experiments, clonogenic assays, and regrowth-delay curves with sensitive and resistant cell lines, we can more accurately evaluate the specificity of the T1 method. Unfortunately, cyclophosphamide-resistant RIF-1 tumors are not available at present. These experiments were, therefore, beyond the scope of the present work and will be presented elsewhere with appropriate tumor models.
This study demonstrates that T1 relaxation is altered in apparently viable regions of the tumor as early as 18 h after cyclophosphamide administration. Significant changes in the T2 relaxation characteristics were not observed during this time period. The observed changes in T1 might be caused by local variations in the macromolecular composition of the tumor after treatment. Elevation in the fractional water content of the extracellular space appears to be an unlikely mechanism because T2 relaxation did not change significantly. The increase in the number of pixels having T1 values greater than threshold occurred in a spatially heterogeneous manner, which might be secondary to vascular compromise induced by cyclophosphamide. These studies suggest that T1 imaging may be useful in evaluating early response to treatment with antitumor therapeutic regimes in the clinical settings.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
1 This work was supported by NIH Grants RR02305 (to J. S. L.), CA51950 (to J. D. G.), and MH11330 (to U. D.). 
2 To whom requests for reprints should be addressed, at B-1 Stellar-Chance Laboratories, 422 Curie Boulevard, Philadelphia, PA 19104. Phone: (215) 898-1805; Fax: (215) 573-2113; E-mail: poptani{at}oasis.rad.upenn.edu
3 Present address: Laboratoire de Biophysique et Traitement de l’Image, 80054 Amines Cedex 1, France.
4 The abbreviations used are: MRI, magnetic resonance imaging; T1, spin-lattice relaxation; T2, spin-spin relaxation; T1, spin-lattice relaxation in the rotating frame; TE, echo time; TSL, spin-lock time; TR, repetition time; hpf, high-power field; TUNEL, terminal deoxynucleotidyl transferase (Tdt)-mediated nick end labeling.
Received 7/14/00. Accepted 8/28/01.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
of protein solutions at very low fields: dependence on molecular weight, concentration, and structure. Magn. Reson. Med., 37: 53-57, 1997.[Medline]
dispersion MRI. Proceedings of the International Society for Magnetic Resonance in Medicine, 2106 ISMRM Berkeley, CA 1999.
. Proceedings of the International Society for Magnetic Resonance in Medicine, 512 ISMRM Berkeley, CA 2000.
and low field 1/T1 of tissue water protons arise from magnetization transfer to macromolecular solid-state broadened lines. Magn. Reson. Med., 28: 145-152, 1992.[Medline]
dispersion imaging and localized T1 dispersion relaxometry: application in vivo to mouse adenocarcinoma. Magn. Reson. Med., 24: 149-157, 1992.[Medline]
This article has been cited by other articles:
|
|
 |
|
  M. I. Kettunen, A. Sierra, M. J. Narvainen, P. K. Valonen, S. Yla-Herttuala, R. A. Kauppinen, and O. H. J. Grohn Low Spin-Lock Field T1 Relaxation in the Rotating Frame as a Sensitive MR Imaging Marker for Gene Therapy Treatment Response in Rat Glioma Radiology, March 1, 2007; 243(3): 796 - 803. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Poptani, N. Bansal, W. T. Jenkins, D. Blessington, A. Mancuso, D. S. Nelson, M. Feldman, E. J. Delikatny, B. Chance, and J. D. Glickson Cyclophosphamide Treatment Modifies Tumor Oxygenation and Glycolytic Rates of RIF-1 Tumors: 13C Magnetic Resonance Spectroscopy, Eppendorf Electrode, and Redox Scanning Cancer Res., December 15, 2003; 63(24): 8813 - 8820. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
O. H. J. Grohn, P. K. Valonen, K. K. Lehtimaki, T. H. Vaisanen, M. I. Kettunen, S. Yla-Herttuala, R. A. Kauppinen, and M. Garwood Novel Magnetic Resonance Imaging Contrasts for Monitoring Response to Gene Therapy in Rat Glioma Cancer Res., November 15, 2003; 63(22): 7571 - 7574. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |
